Predictive Validity of UTME and Post-UTME Chemistry Scores as Joint Predictors of First-Year Achievement in Chemistry-Based Courses in Enugu State Public Universities

Abstract

This study investigated the predictive validity of UTME and Post-UTME Chemistry scores as joint predictors of first-year achievement in chemistry-based courses in Enugu State public universities. A predictive correlational survey design was adopted. The population was 2,396 students admitted in 2007/2008 and 2008/2009 academic sessions into Agricultural Economics and Biochemistry in University of Nigeria, Nsukka (Federal) and Enugu State University of Science and Technology (State). Purposive sampling was used to select 349 students with complete records (UNN = 190, ESUT = 159). A broadsheet proforma was used for data collection. Data were analyzed using Pearson correlation and Multiple Linear Regression at 0.05 alpha. Results showed low to moderate inter-correlation between UTME and Post-UTME (r = 0.01 - 0.48) indicating no serious multicollinearity. For Federal university, UTME and Post-UTME jointly predicted first-year achievement significantly; Agricultural Economics: Multiple r = 0.58, R² = 0.34, F = 35.42, p < .001; Biochemistry: r = 0.51, R² = 0.26, p < .001. For State university, joint prediction was lower; Agricultural Economics: r = 0.41, R² = 0.17, p = .001; Biochemistry: r = 0.38, R² = 0.14, p = .055. Combined model explained 27% variance, F(2, 346) = 64.18, p < .001. Post-UTME had higher standardized Beta (β = 0.40, p < .001) than UTME (β = 0.17, p = .003). It was concluded that UTME and Post-UTME Chemistry have moderate predictive validity in Federal but low in State university, suggesting institutional differences in Post-UTME standard. Recommendation was made to retain Post-UTME while standardizing its conduct in State universities.